Metraipolis › archive › dc_6989583b71b5
ProCraft: Instruction-Driven Music Production from Free-Form Text — arXiv:2611.07423
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.07423
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T15:00:08.854Z
- SHA-256 of the archived text
f265995a3f4ee8346f4f7e581dc42b2fb6de10489f887079a1e5d2dcae88fb85- Stable URL of this page
https://metraipolis.com/documento/dc_6989583b71b5
How to cite
ProCraft: Instruction-Driven Music Production from Free-Form Text — arXiv:2611.07423 — arXiv (en), https://arxiv.org/abs/2611.07423 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6989583b71b5 (Metraipolis archive dc_6989583b71b5, sha256 f265995a3f4ee8346f4f7e581dc42b2fb6de10489f887079a1e5d2dcae88fb85)
BibTeX
@misc{dc6989583b71b5,
title = {ProCraft: Instruction-Driven Music Production from Free-Form Text — arXiv:2611.07423},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.07423},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6989583b71b5; sha256 f265995a3f4ee8346f4f7e581dc42b2fb6de10489f887079a1e5d2dcae88fb85},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07423. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Music production typically involves multiple editing operations that each require specialized tools and expertise. Existing text-guided music editing systems support few operations or instruments, operate on short clips, or handle only one operation per instruction, unlike real workflows where several changes are requested at once. Here, we present ProCraft, a lightweight 0.5B-parameter autoregressive codec language model that edits 30s music excerpts end-to-end from free-form instructions, including compound requests. ProCraft supports seven operations across eight instruments, from tasks with a unique target (e.g., track removal) to creative ones (e.g., accompaniment generation). We align intermediate representations with pretrained music embeddings and pitch and instrument targets to preserve content outside the edit, and gated cross-attention in every decoder layer to keep generation tied to the instruction. To train on compound requests, we introduce On-the-fly Editing-pair Synthesis, which generates paired instructions and audio combining track-level operations with effect changes. ProCraft performs competitively against task-specific systems on single-task editing, and outperforms four free-text editors on compound instructions for track removal, extraction and denoising on real (MoisesDB) and synthetic (Slakh) recordings. In a user study (16 listeners), ProCraft was rated highest on removal and denoising and on-par with a 16-times larger model on extraction and instrument transfer. Ablations further show that representation alignment and cross-attention reduce leakage from unwanted tracks and raise editing success rates, especially on out-of-domain data. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv eess.AS, nuovi lavori del giorno). Categorie arXiv: Sound (cs.SD) ; Audio and Speech Processing (eess.AS). Pagina del lavoro: https://arxiv.org/abs/2611.07423 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-06T14:42:06.075Z dalla pagina https://arxiv.org/list/eess.AS/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).